论文部分内容阅读
风电集群点出力预测是提高大规模风电接入条件下电力系统运行安全性和经济性的有效手段。针对风电集群点出力的模式性，利用元组向量表示其变化趋势，通过元组向量时间扭曲法对当前出力模式在历史出力模式库中进行匹配，以时间偏移量、时间压缩率、幅值压缩率3个修正参数表征相似出力模式的差异性，结合匹配结果对风电集群点出力进行预测。实际应用结果表明，该方法可以取得较为理想的效果，参数修正能够提高预测精度，使预测结果更加合理。“,”ABSTRACT:Power prediction of wind farm cluster is an effective means to improve the operational security and economy of power system integrated with large-scale wind power. Based on the output profile pattern of wind farm cluster, this paper adopted the tuple vector to represent output variation trend, matched latest output pattern among history database by the method of tuple vector time warp (TVTW), characterized differences between similar output patterns by time offset, time and amplitude compression ratios, and predicted clustered wind power based on matching results finally. The practical system application shows that the proposed method is able to obtain satisfactory prediction effects, which could be more reasonable by parameters correcting.